Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Boston Dynamics Atlas Robot Powered by AI "Large Behavior Models" Boston Dynamics and Toyota just showed Atlas doing something wild, packing boxes using a Large Behavior Model. One AI brain controls walking, crouching, lifting, everything. Just learned from human demos.

Robots Digest 🤖

4,830 subscribers

25,305 views • 8 months ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Boston Dynamics and Toyota Research Institute are advancing humanoid robotics with Large Behavior Models (LBMs) for Atlas.

Boston Dynamics and Toyota Research Institute are advancing humanoid robotics with Large Behavior Models (LBMs) for Atlas.

The Humanoid Hub

123,893 views • 11 months ago

Boston Dynamics has released a new video of their next-generation humanoid robot, Atlas. "We are excited to share some of our progress on developing Large Behavior Models for Atlas. This work is part of a collaboration between AI research teams at Toyota Research Institute (TRI) and Boston Dynamics. We have been building end-to-end language-conditioned policies that enable Atlas to accomplish long-horizon manipulation tasks."

Boston Dynamics has released a new video of their next-generation humanoid robot, Atlas. "We are excited to share some of our progress on developing Large Behavior Models for Atlas. This work is part of a collaboration between AI research teams at Toyota Research Institute (TRI) and Boston Dynamics. We have been building end-to-end language-conditioned policies that enable Atlas to accomplish long-horizon manipulation tasks."

Sawyer Merritt

200,744 views • 11 months ago

The Large Behavior Model (LBM), co-developed by Boston Dynamics and TRI, can perform various manipulation tasks on Atlas MTS (the Atlas upper-body robot). The training process is the same whether it’s manipulating a 22 lb car tire, tying a rope, or flipping a barstool.

The Large Behavior Model (LBM), co-developed by Boston Dynamics and TRI, can perform various manipulation tasks on Atlas MTS (the Atlas upper-body robot). The training process is the same whether it’s manipulating a 22 lb car tire, tying a rope, or flipping a barstool.

The Humanoid Hub

40,520 views • 11 months ago

When 60 Minutes visited Boston Dynamics in 2021, Atlas was a bulky, hydraulic robot. Back then, it relied on algorithms written by engineers. Today, Atlas has a sleek, all-electric body and an AI-powered brain, smart enough to pull off hard-to-believe feats autonomously.

When 60 Minutes visited Boston Dynamics in 2021, Atlas was a bulky, hydraulic robot. Back then, it relied on algorithms written by engineers. Today, Atlas has a sleek, all-electric body and an AI-powered brain, smart enough to pull off hard-to-believe feats autonomously.

60 Minutes

62,774 views • 7 months ago

Boston Dynamics taught Atlas a Rabona kick. Learned from human mocap, retargeted to Atlas, trained in sim via RL, deployed zero-shot to the real robot. Soccer skills demand whole-body coordination, and similar recipe transfers to warehouse work.

Boston Dynamics taught Atlas a Rabona kick. Learned from human mocap, retargeted to Atlas, trained in sim via RL, deployed zero-shot to the real robot. Soccer skills demand whole-body coordination, and similar recipe transfers to warehouse work.

The Humanoid Hub

11,268 views • 2 months ago

Boston Dynamics gave a glimpse at the next version of its Atlas robot at #CES2026, and its predecessor made up for its lack of movement. And now, we know they'll be powered by Google DeepMind AI:

Boston Dynamics gave a glimpse at the next version of its Atlas robot at #CES2026, and its predecessor made up for its lack of movement. And now, we know they'll be powered by Google DeepMind AI:

TechCrunch

44,836 views • 7 months ago

Boston Dynamics has demonstrated its new Atlas humanoid robot lifting and carrying a 50-pound fridge. The robot uses its whole body to balance and move the heavy object in a natural, human-like way. Before doing it in the real world, Atlas practiced the task for millions of hours in a virtual simulation.

Boston Dynamics has demonstrated its new Atlas humanoid robot lifting and carrying a 50-pound fridge. The robot uses its whole body to balance and move the heavy object in a natural, human-like way. Before doing it in the real world, Atlas practiced the task for millions of hours in a virtual simulation.

Space and Technology

180,369 views • 1 month ago

🚨 BREAKING: Hyundai Worldwide to deploy tens of thousands of robots! Hyundai is scaling up its robotics strategy with a major investment in Boston Dynamics ' robotic platforms. The company plans to deploy tens of thousands of units — including the Atlas humanoid, Spot quadruped, and Stretch trailer-unloading robots—across its manufacturing and logistics operations in the coming years. Spot is already in use at Hyundai’s facilities, performing industrial inspection and predictive maintenance tasks. Atlas, Boston Dynamics’ bipedal robot, is next in line for factory deployment, aimed at addressing physically demanding and repetitive jobs. Beyond deployment, Hyundai will support Boston Dynamics’ growth by integrating the automaker’s manufacturing capabilities to boost robot production. The move is part of Hyundai’s broader ambition to lead in robotics and “physical AI”. That's so cool to see the scale-up phase with humanoids. Looking forward to seeing Atlas live in Boston this month. 🔥

🚨 BREAKING: Hyundai Worldwide to deploy tens of thousands of robots! Hyundai is scaling up its robotics strategy with a major investment in Boston Dynamics ' robotic platforms. The company plans to deploy tens of thousands of units — including the Atlas humanoid, Spot quadruped, and Stretch trailer-unloading robots—across its manufacturing and logistics operations in the coming years. Spot is already in use at Hyundai’s facilities, performing industrial inspection and predictive maintenance tasks. Atlas, Boston Dynamics’ bipedal robot, is next in line for factory deployment, aimed at addressing physically demanding and repetitive jobs. Beyond deployment, Hyundai will support Boston Dynamics’ growth by integrating the automaker’s manufacturing capabilities to boost robot production. The move is part of Hyundai’s broader ambition to lead in robotics and “physical AI”. That's so cool to see the scale-up phase with humanoids. Looking forward to seeing Atlas live in Boston this month. 🔥

Lukas Ziegler

64,266 views • 1 year ago

.Mario Bollini leads the Atlas product team at Boston Dynamics. Mario sees humanoid robots as the next-level tools embodying generalizable AI that could deliver meaningful commercial value. We discuss Mario’s journey in robotics, product lessons from Spot applied to Atlas, the hardware approach for Atlas, AI, potential commercial applications, home robots, the competition from China, and more. 00:17 Personal journey into robotics 02:20 Learnings from Spot 04:09 Atlas actuation - hydraulic to electric 05:36 Evolution of Atlas' movement control 07:55 Actuator design trade-offs 09:35 Design decision for Atlas' leg rotation 11:21 Stance on humanoid form factor 13:13 Customer skepticism; adoption hurdles 14:49 Atlas applications; current pilots 17:14 Atlas' readiness for human collaboration 19:46 Why not a five-finger hand 21:10 Explaining Large Behavior Models 22:26 Why is RAI a separate entity 24:04 Chicken-and-egg problem of robotics 27:09 Home use humanoids 29:14 Preventing weaponization 31:00 Competition from China 32:13 Long term future vision 33:55 Reason for Atlas's round head 35:27 Any final thoughts?

.Mario Bollini leads the Atlas product team at Boston Dynamics. Mario sees humanoid robots as the next-level tools embodying generalizable AI that could deliver meaningful commercial value. We discuss Mario’s journey in robotics, product lessons from Spot applied to Atlas, the hardware approach for Atlas, AI, potential commercial applications, home robots, the competition from China, and more. 00:17 Personal journey into robotics 02:20 Learnings from Spot 04:09 Atlas actuation - hydraulic to electric 05:36 Evolution of Atlas' movement control 07:55 Actuator design trade-offs 09:35 Design decision for Atlas' leg rotation 11:21 Stance on humanoid form factor 13:13 Customer skepticism; adoption hurdles 14:49 Atlas applications; current pilots 17:14 Atlas' readiness for human collaboration 19:46 Why not a five-finger hand 21:10 Explaining Large Behavior Models 22:26 Why is RAI a separate entity 24:04 Chicken-and-egg problem of robotics 27:09 Home use humanoids 29:14 Preventing weaponization 31:00 Competition from China 32:13 Long term future vision 33:55 Reason for Atlas's round head 35:27 Any final thoughts?

The Humanoid Hub

21,372 views • 9 months ago

Scientists developed HumanX, a system that lets humanoid robots learn new skills by watching just one short human video. It learns directly from demonstrations instead of large datasets or complex training rewards. Using this, the Unitree G1 humanoid robot learned 10 different real-world skills without extra training.

Scientists developed HumanX, a system that lets humanoid robots learn new skills by watching just one short human video. It learns directly from demonstrations instead of large datasets or complex training rewards. Using this, the Unitree G1 humanoid robot learned 10 different real-world skills without extra training.

Space and Technology

27,686 views • 3 months ago

$Boston Dynamics’ Spot costs $75,000. This one? $135. LeCabot is an open-source mod for your SO-100 and Unitree Go2 that turns them into a fully capable mobile manipulator… for a fraction of the price. You control both the robot dog and arm simultaneously using a Meta Quest headset, thanks to Phospho AI 🧪 It’s real, it’s accessible, and it’s ready for AI model training on rough terrain. Start building: Bill of materials and assembly instructions →$

Boston Dynamics’ Spot costs $75,000. This one? $135. LeCabot is an open-source mod for your SO-100 and Unitree Go2 that turns them into a fully capable mobile manipulator… for a fraction of the price. You control both the robot dog and arm simultaneously using a Meta Quest headset, thanks to Phospho AI 🧪 It’s real, it’s accessible, and it’s ready for AI model training on rough terrain. Start building: Bill of materials and assembly instructions →

Ilir Aliu - eu/acc

20,614 views • 11 months ago

Boston Dynamics just gave its robot dog a brain that reasons about the physical world. Google DeepMind's Gemini Robotics model is now running inside Spot, the four-legged robot already deployed at thousands of facilities worldwide. The upgrade centers on something called embodied reasoning. Basically, instead of following rigid instructions, Spot now interprets what it sees and makes decisions on its own. One standout example: Spot can now look at a pressure gauge, zoom in on the needle, calculate the intervals between tick marks, and deliver an accurate reading. All autonomously. The model also introduces: > Multi-view reasoning, combining feeds from multiple cameras to understand its environment even when objects are blocked from one angle > Built-in success detection, so Spot knows when it's failed and decides whether to retry or move on That last part is a key requirement for true autonomy. Most companies building AI-powered robots are still stuck in the demo phase. Boston Dynamics is one of the few with actual paying customers... which means thousands of Spots collecting real-world data that feeds back into making the AI smarter. That flywheel is hard to compete with.

Boston Dynamics just gave its robot dog a brain that reasons about the physical world. Google DeepMind's Gemini Robotics model is now running inside Spot, the four-legged robot already deployed at thousands of facilities worldwide. The upgrade centers on something called embodied reasoning. Basically, instead of following rigid instructions, Spot now interprets what it sees and makes decisions on its own. One standout example: Spot can now look at a pressure gauge, zoom in on the needle, calculate the intervals between tick marks, and deliver an accurate reading. All autonomously. The model also introduces: > Multi-view reasoning, combining feeds from multiple cameras to understand its environment even when objects are blocked from one angle > Built-in success detection, so Spot knows when it's failed and decides whether to retry or move on That last part is a key requirement for true autonomy. Most companies building AI-powered robots are still stuck in the demo phase. Boston Dynamics is one of the few with actual paying customers... which means thousands of Spots collecting real-world data that feeds back into making the AI smarter. That flywheel is hard to compete with.

Rowan Cheung

54,376 views • 3 months ago

Boston Dynamics just dropped the production-ready Atlas at CES 2026, and the specs are next-level. This isn't just a prototype; it's a rugged, enterprise-grade worker built for the grind of real-world factories.🤖🦾 With 56 degrees of freedom and fully rotational joints, Atlas moves with "superhuman" flexibility. It features human-scale hands with tactile sensing in every finger and palm, allowing for high-precision manipulation. Plus, its 360-degree vision system ensures it always knows who's approaching, keeping human-robot collaboration safe. ➤ Rugged Performance: Lifts up to 110 lbs (approx. 50kg) and operates in extreme temps from -4°F to 104°F (-20°C to 40°C). It’s even water-resistant for industrial wash-downs. ➤ Smart Endurance: Runs for 4 hours on dual swappable batteries. When power gets low, it autonomously heads back to the station to swap its own batteries and get right back to work. ➤ Cloud Intelligence: New tasks can be trained in less than a day. Thanks to the Orbit platform, once one Atlas learns a skill, it’s instantly shared across the entire fleet. The scale is massive: the entire 2026 supply is already spoken for by Hyundai and a new AI partner. Hyundai is also building a specialized "Robotics Metal Plant" and a new factory capable of churning out 30,000 Atlas units annually. We’re officially watching the birth of a humanoid workforce at scale. Source: Boston Dynamics & Hyundai Group

Boston Dynamics just dropped the production-ready Atlas at CES 2026, and the specs are next-level. This isn't just a prototype; it's a rugged, enterprise-grade worker built for the grind of real-world factories.🤖🦾 With 56 degrees of freedom and fully rotational joints, Atlas moves with "superhuman" flexibility. It features human-scale hands with tactile sensing in every finger and palm, allowing for high-precision manipulation. Plus, its 360-degree vision system ensures it always knows who's approaching, keeping human-robot collaboration safe. ➤ Rugged Performance: Lifts up to 110 lbs (approx. 50kg) and operates in extreme temps from -4°F to 104°F (-20°C to 40°C). It’s even water-resistant for industrial wash-downs. ➤ Smart Endurance: Runs for 4 hours on dual swappable batteries. When power gets low, it autonomously heads back to the station to swap its own batteries and get right back to work. ➤ Cloud Intelligence: New tasks can be trained in less than a day. Thanks to the Orbit platform, once one Atlas learns a skill, it’s instantly shared across the entire fleet. The scale is massive: the entire 2026 supply is already spoken for by Hyundai and a new AI partner. Hyundai is also building a specialized "Robotics Metal Plant" and a new factory capable of churning out 30,000 Atlas units annually. We’re officially watching the birth of a humanoid workforce at scale. Source: Boston Dynamics & Hyundai Group

RoboHub🤖

28,554 views • 7 months ago

Humanoid robotics is entering a new phase： Xiaomi Robotics 0 (XR-0) explores learning manipulation directly from large-scale human videos — aligning video understanding with robot embodiment to generate executable actions. Watch humans->learn representations ->map to robot control. Robot data is scarce. Internet video is not. If video pretraining transfers reliably into embodied control, the scaling law of robotics changes. Deployment speed becomes a function of model alignment, not just hardware iteration. We’re moving from engineered skills to learned behavior. That’s a structural shift. project:

Humanoid robotics is entering a new phase： Xiaomi Robotics 0 (XR-0) explores learning manipulation directly from large-scale human videos — aligning video understanding with robot embodiment to generate executable actions. Watch humans->learn representations ->map to robot control. Robot data is scarce. Internet video is not. If video pretraining transfers reliably into embodied control, the scaling law of robotics changes. Deployment speed becomes a function of model alignment, not just hardware iteration. We’re moving from engineered skills to learned behavior. That’s a structural shift. project:

CyberRobo

18,356 views • 5 months ago

There's a fruit fly walking around right now that was never born. eon just released a video where they took a real fly's connectome — the wiring diagram of its brain — and simulated it. Dropped it into a virtual body. It started walking. Grooming. Feeding. Doing what flies do. Nobody taught it to walk. No training data, no gradient descent toward fly-like behavior. This is the opposite of how AI works. They rebuilt the mind from the inside, neuron by neuron, and behavior just... emerged. It's the first time a biological organism has been recreated not by modeling what it does, but by modeling what it is. A human brain is 6 OOM more neurons. That's a scaling problem, something we've gotten very good at solving. So what happens when we have a working copy of the human mind?

There's a fruit fly walking around right now that was never born. eon just released a video where they took a real fly's connectome — the wiring diagram of its brain — and simulated it. Dropped it into a virtual body. It started walking. Grooming. Feeding. Doing what flies do. Nobody taught it to walk. No training data, no gradient descent toward fly-like behavior. This is the opposite of how AI works. They rebuilt the mind from the inside, neuron by neuron, and behavior just... emerged. It's the first time a biological organism has been recreated not by modeling what it does, but by modeling what it is. A human brain is 6 OOM more neurons. That's a scaling problem, something we've gotten very good at solving. So what happens when we have a working copy of the human mind?

hattie

9,361,810 views • 5 months ago

Tired of teleoperation? One human video → 1,000s of robot demos. (📍GitHub ) Scaling Robot Data Without Dynamics Simulation or Robot Hardware Real2Render2Real (R2R2R) is a new way to scale robot data without physics simulation or hardware. You take a phone scan + a single monocular human demo. It tracks the motion, renders photorealistic scenes, and generates diverse, robot-agnostic trajectories ready for training. > No teleop, no sim, no robot, just a phone and a video > Train VLA models and diffusion policies directly on the output > Supports multiple robot embodiments with kinematic consistency > 1000s of demos in 1/27 the time of real-world collection Thank you, Max Fu, for sharing!! Project: Paper: Code coming soon: It shows that with the right pipeline, you can scale robot learning data without touching a robot. One of the most interesting directions in scalable robotics today. —— Weekly robotics and AI insights. Subscribe free:

Tired of teleoperation? One human video → 1,000s of robot demos. (📍GitHub ) Scaling Robot Data Without Dynamics Simulation or Robot Hardware Real2Render2Real (R2R2R) is a new way to scale robot data without physics simulation or hardware. You take a phone scan + a single monocular human demo. It tracks the motion, renders photorealistic scenes, and generates diverse, robot-agnostic trajectories ready for training. > No teleop, no sim, no robot, just a phone and a video > Train VLA models and diffusion policies directly on the output > Supports multiple robot embodiments with kinematic consistency > 1000s of demos in 1/27 the time of real-world collection Thank you, Max Fu, for sharing!! Project: Paper: Code coming soon: It shows that with the right pipeline, you can scale robot learning data without touching a robot. One of the most interesting directions in scalable robotics today. —— Weekly robotics and AI insights. Subscribe free:

Ilir Aliu

42,864 views • 6 months ago

$Boston Dynamics’ Spot costs $75,000. This one? $135. LeCabot is an open-source mod for your SO-100 and Unitree Go2 that turns them into a fully capable mobile manipulator… for a fraction of the price. You control both the robot dog and arm simultaneously using a Meta Quest headset, thanks to phospho 🧪 🧪 It’s real, it’s accessible, and it’s ready for AI model training on rough terrain. Thanks for sharing, ulyanov! Start building: Bill of materials and assembly instructions →$

Boston Dynamics’ Spot costs $75,000. This one? $135. LeCabot is an open-source mod for your SO-100 and Unitree Go2 that turns them into a fully capable mobile manipulator… for a fraction of the price. You control both the robot dog and arm simultaneously using a Meta Quest headset, thanks to phospho 🧪 🧪 It’s real, it’s accessible, and it’s ready for AI model training on rough terrain. Thanks for sharing, ulyanov! Start building: Bill of materials and assembly instructions →

Ilir Aliu - eu/acc

21,874 views • 1 year ago

NEWS: Boston Dynamics has just released a new video of its upgraded next-generation humanoid robot called Atlas. • 4 hour battery. Self-swappable for continuous operation • 6 feet 2 inches tall • Weight: 198 lbs • 56 total degrees of freedom • Now fully electric, ditching older hydraulic systems • New lightweight mix of aluminum and titanium components • 110 lbs weight capacity (66 lbs sustained) • Can reach up to 7.5 ft • Constantly evaluates its surroundings and adjusts its posture, balance, and grip in real time • Hands that can reconfigure as needed. Tactile sensors feed data back into the system, helping apply the right amount of force • Brain is powered by Nvidia chips

NEWS: Boston Dynamics has just released a new video of its upgraded next-generation humanoid robot called Atlas. • 4 hour battery. Self-swappable for continuous operation • 6 feet 2 inches tall • Weight: 198 lbs • 56 total degrees of freedom • Now fully electric, ditching older hydraulic systems • New lightweight mix of aluminum and titanium components • 110 lbs weight capacity (66 lbs sustained) • Can reach up to 7.5 ft • Constantly evaluates its surroundings and adjusts its posture, balance, and grip in real time • Hands that can reconfigure as needed. Tactile sensors feed data back into the system, helping apply the right amount of force • Brain is powered by Nvidia chips

Sawyer Merritt

2,441,004 views • 7 months ago